Slavery in Ancient Greece

Abstract

Slavery was a defining phenomenon in the history of Greek city-states. This chapter explores the various local slaving systems of ancient Greece and examines their similarities and differences with respect to four factors: the slaving strategies employed by the masters; the geopolitical context of ancient Greek communities; the ways and aims for which the political community intervened in the master–slavery relationship; and the forms of communities in which slaves participated.

Introduction

Slavery in the Greek city-states has held a foundational role in the modern study of global slavery. Moses Finley, an influential ancient historian, coined the famous distinction between “societies with slaves,” where slaves are few and slavery plays a limited economic role, and “slave societies,” where slavery constitutes a dominant economic, social, and political institution and slaves comprise a substantial proportion of the population. Finley argued that while societies with slaves have been ubiquitous in global history, there had been only five slave societies: Greece, Rome, the US South, Brazil, and the Caribbean. In this approach, Greek city-states had the “honor” of being the first slave societies in world history.¹

The traditional approach to Greek slavery has been based on two major tenets. The first is that early Greek communities were originally societies with slaves, where the dependent labor of the free lower classes was the main source of elite wealth. But in the course of the archaic period (700–500 BCE), the lower classes gained citizenship rights and could no longer be directly exploited by Greek elites, who turned to the mass importation of slaves, leading to the emergence of slave societies. The second tenet is the assumption that Greek slavery is tantamount to Athenian slavery in the classical period (500–300 BCE), from where most of our evidence comes; the other Greek cities are assumed to have essentially the same slave system as Athens. This idea was further supported by the assumption that servile groups like the helots of Sparta and the woikeis of Crete were not “proper” slaves, but should rather be seen as akin to medieval serfs, since they lived in family groups as dependent peasants. According to the traditional approach, slavery in the Greek world consisted of those slave systems that resemble the familiar image of slavery in the US South; any system that diverged significantly can be explained away as not being “proper” slavery.²

This chapter is based on the revolutionary implications of recent research over the last decade, which has seriously challenged the major assumptions of the traditional approach. A careful reading of the Homeric and Hesiodic epics, our earliest sources for Greek history, has revealed that the elites in the Homeric world (800–700 BCE) depended overwhelmingly on slave labor; accordingly, early Greek communities were already slave societies. We do not know how far back slavery was a dominant phenomenon in the Greek world, but it is obvious that the traditional narrative of a transition from societies with slaves into slave societies in the course of the archaic period is no longer tenable. We need a new kind of narrative to explain the differences between the forms of slavery attested in different historical periods. At the same time, scholars have started to accept that Greek slavery was not a uniform phenomenon, but consisted of various local slave systems, each of which had developed its own peculiar features. Spartan helots, for example, were not serfs, but slaves with peculiar characteristics as a result of the particular historical development of Spartan society. Slavery does not have some trans-historical essence, but is the historical outcome of the interplay between strategies employing human property for various ends and the wider processes and contexts within which these strategies take place.³

Instead of the traditional narrative and interpretative framework, we should rather locate the history of Greek slave systems within a number of processes. We will focus on four distinct but interrelated historical processes. Prime of place goes to the process of growing connectivity that from the archaic period onward came to interlink various areas of the Mediterranean and the Black Sea; this process was partly based on decentralized networks moving goods, people, ideas and technologies, and partly on attempts by states and potentates to canalize connectivity for their own ends. Greek slave systems cannot be understood outside this quantum leap in the connectivity of the Mediterranean and the Black Sea during the first millennium BCE; in the same way that early modern slavery is incomprehensible outside the emergence of the Atlantic world that interlinked European, African, Native American and Colonial American societies, economies and cultures, Greek slave systems were intimately related to other Mediterranean and Black Sea slave systems. Increasing connectivity set the stage for drastic changes in Greek material culture; it made possible the utilization of Mediterranean micro-ecological diversity and fragmentation through large-scale processes of exchange and redistribution. The resulting specialization, production for the market, dependence on exchange and surplus accumulation went hand in hand with the emergence of the first consumer societies, in which substantial social strata desired and consumed goods from various areas of the Mediterranean and the Black Sea. Slavery was deeply inscribed in this process, not only in terms of the slave trade, but also in terms of producing a major part of these various goods and creating the surpluses that allowed the emergence of consumer societies.

At the same time, processes of community formation and claim-making transformed the socio-political settings of Greek city-states. This process shaped the institutions of Greek city-states and the meaning of citizenship and changed the ways in which Greek communities formulated the distinction between insiders and outsiders. Freedom was no longer the status of not being property, and started to acquire additional features that tended to turn into a total and unalterable status; freeborn people could no longer lose their status within their community and their status protected them from dishonor and physical punishment. The formalization of free and slave status created major disadvantages for slaves, but at the same time opened new institutional settings that slaves could potentially take advantage of. Political communities could also superimpose their own priorities on masters and slaves, limiting what could be done to slaves or what slaves could do. Finally, geopolitical processes redefined how violence and ideology affected enslavement and liberation across the Mediterranean and the Black Sea. Most slaves were produced through warfare, raiding and international trade. This means that slavery was directly inscribed in the changing history of the forms of warfare, predation, exchange, state-building and empire-building that linked together communities into wider systems of international relations. The emergence of large states and empires in various parts of the Mediterranean and the Black Sea, as well as their occasional collapse, enhanced the scale and stakes of warfare and the extent of slave-making, creating large and interconnected slaving zones; at the same time, the peculiar form of the Greek geopolitical system had also important implications for the emergence of various forms of no-slaving zones.⁴

It might be worth offering a general typology of Greek slave systems at this point. The first group of Greek slave systems comprised societies like Sparta and Crete, whose citizenries consisted of leisured gentlemen exploiting slave labor. These systems had limited engagement with Mediterranean connectivity, being agricultural societies geared toward local production and consumption. Slaves constituted the majority population group in these societies, and their replenishment was based on natural reproduction. A second group of slave systems consisted of relatively wealthy societies with deep engagement with Mediterranean connectivity, highly diversified economies and often a predilection for marketable crops, like oil and wine. Apart from Athens, which was in addition an imperial center, this group included primarily island and coastal communities like Chios, Corfu and Aegina. These slave systems depended largely on trade for the replenishment of their slave populations. The historian Thucydides believed that Chios had more slaves than any other Greek city, even Sparta, presumably as a proportion of the population rather than an absolute number⁵; it is possible therefore that some Greek cities like Chios had slave populations that approached those in some early modern Caribbean colonies. In the case of Athens, most scholars would accept a guesstimate between 20 and 50 percent of the population. The third group, which comprised the majority of Greek societies, was similar to the second, but with far fewer resources and extent of connectivity; we should therefore expect a social structure similar to the second group, but with fewer slaves given their limited wealth; slave populations of 20 percent or less should probably be expected.⁶

Entry

Cross-culturally, we can distinguish between four major forms of entry into slavery: (a) internal enslavement within a community; (b) violent enslavement through state warfare or piracy; (c) the slave trade and (d) the inheritance of slave status through the natural reproduction of slave populations. As we shall see, we can find both common patterns across the Greek world, as well as major differences.

One of the peculiar features of Greek slave systems is the limited role of internal forms of enslavement. Solon’s reforms in early sixth-century BCE Athens prohibited debt slavery; Athenians could no longer be enslaved for debt within their community. We have no concrete evidence about most other Greek communities, but we get the impression that enslavement for debt was marginal, if not equally prohibited. On the other hand, the existence of debt bondage is attested; free people had to work for their lenders in order to repay their debts in conditions that were often akin to slavery, although they retained their free status while in debt bondage. Largely invisible is the right of fathers to sell their children, as was the case in many other ancient societies. Finally, penal enslavement is unattested for citizens, although we know that in Athens it was a possible punishment for free foreign residents who had not paid their special taxes, or for people who attempted to usurp the right to citizenship. Internal forms of enslavement were not unknown in Greek societies, but as a result of community protection of the free status of citizens they were marginal phenomena in the Greek world.⁷

Greek city-states had very strong no-slaving zones, but these zones concerned only their own citizens; the rest of the world, including citizens of other Greek city-states, were considered potentially enslaveable. The slaving zones of Greek city-states were therefore enormous; not only could Greek city-states potentially enslave their neighbors, but, due to the connectivity expansion we examined above, they could also receive slaves from areas like Asia Minor, the Levant, Thrace and the Black Sea. Given the limited role of internal forms of enslavement, it was violence and trade that constituted the main forms of entry into slavery in the Greek world.

Generally speaking, in all periods of Greek history it was taken for granted that violence was a legitimate means of generating slaves; but this general fact was often hedged in by certain important qualifications. The first important qualification is the distinction between oligopolistic and monopolistic states in terms of their recourse to violence; oligopolistic states allowed their citizens to employ violence against foreigners for their private gain, and required their citizens to contribute their military equipment and ships only on those occasions in which the state was fighting against another state; monopolistic states restricted the use of violence to state purposes only, prohibiting or discouraging their citizens from using violence for private gain, and often maintaining substantial state arsenals.⁸ Until the late archaic period, all Greek states were oligopolistic; accordingly, elites and commoners habitually engaged in piracy and other forms of violence that produced movable wealth and captives. The famous poem of Hybrias presents a Cretan master who attributes his wealth, leisure and cowering slaves to his military prowess.⁹

From the late archaic period onward, though, many Greek states made the transition from oligopolistic to monopolistic forms: although slaves were still produced by state warfare, their citizens could no longer engage in acts of private enslavement through violence. In some parts of the Greek world, like Crete and Aetolia, states remained oligopolistic down to the end of the Hellenistic period (323–31 BCE); piracy remained an important form of enslavement in these areas, and, depending on circumstances, it could occasionally contribute substantial numbers of slaves. But in the main areas of the Greek world (the Peloponnese, central Greece, the Aegean islands and coastal Asia Minor), the emergence of monopolistic states made private violent enslavement by elites and commoners a phenomenon of the past.¹⁰

What was the role of state violence in producing slaves in these areas? The second important qualification is the limited role of transcultural wars in Greek history, a major aspect of global enslavement. There clearly existed military conflicts between (some) Greeks and (some) non-Greeks, and the Persian Wars (490–478 BCE) are the most famous example of them; but, by and large, most episodes of warfare in the Greek world involved Greeks fighting against other Greeks. Given this fact, the large numbers of non-Greek slaves in Greek city-states cannot be attributed to warfare between Greeks and non-Greeks; not because Greeks would not have enslaved their non-Greek enemies in large numbers, but because they rarely had the opportunity to do so. Accordingly, enslavement by war in the Greek world is effectively tantamount to the enslavement of Greeks by other Greeks.

As far as the archaic period is concerned, the enslavement of defeated enemies appears to be a relatively common phenomenon. According to much later accounts, the helots were populations conquered and enslaved by Sparta, while similar narratives exist for the conquest and enslavement of the Penestai of Thessaly or the Mariandynoi in Heraclea Pontica.¹¹ Whether these accounts are historically reliable is debated by modern historians; but once we reach the classical period, for which we are much better informed, we encounter a paradox. On the one hand, the Greeks considered perfectly legitimate the enslavement of their Greek opponents: among many actual examples, we can mention the enslavement of the Melians by the Athenians and the enslavement of the Thebans by Alexander the Great. On the other hand, given the ubiquity of warfare in the Greek world, the enslavement of defeated opponents appears as a relatively rare outcome of the fate of captives. Furthermore, Greek texts give the impression that slaves in Greek cities were almost exclusively non-Greek, despite the extant evidence for the enslavement of Greeks by other Greeks.

How should we explain this paradox? Greek city-states could deal with defeated opponents and captives in a variety of ways: they could exchange prisoners, release the free captives and keep those who were already slaves, use the captives as bargain chips for a wider settlement, kill all captives, kill the male adults and enslave the rest, or enslave all captives. Enslavement of captives was a form of conspicuous destruction: it was undertaken as a public statement, whether as revenge for heinous crimes, to discourage future resistance or to exterminate an enemy community. Accordingly, the enslavement of Greek opponents was only undertaken in particular circumstances and for specific purposes, rather than as a default policy. The evolution of Greek interstate relations was also an important factor: the creation of hegemonic alliances and the development of means of incorporating defeated communities in state structures offered alternatives to enslavement for victorious states. Finally, Greek city-states developed robust ransoming mechanisms, by signing multilateral treaties or encouraging individuals to ransom fellow citizens or friendly foreigners.¹²

As a result of all these factors, while the enslavement of Greeks in war remained a constant factor of entry into slavery, it was rather trade that constituted the main source of slaves in most areas of the Greek world. The processes of connectivity that we mentioned above ensured that Greek communities could draw slaves from all areas of the eastern Mediterranean and the Black Sea. This constituted an enormous expansion of the reach of slaving zones in the ancient world. Geopolitical changes in various parts of the Mediterranean and the Black Sea increased significantly the supply of slaves. The warfare that accompanied the creation of large states and empires, like the Odrysian kingdom in Thrace or the kingdom of the Royal Scythians in the Black Sea, provided major opportunities for slave-making; the same applied to the crisis or collapse of such states. The Greek colonies in the Western Mediterranean, Thrace and the Black Sea and their commercial networks provided a major, easy and profitable outlet for the thousands of slaves generated by these conflicts. A single day of campaigning by Seuthes, a Thracian potentate around 400 BCE who was trying to extend his authority, produced a thousand captives that were quickly disposed of in the nearby Greek colony of Perinthos.¹³

Furthermore, the maritime character of Mediterranean connectivity offered significant advantages in relation to the land-based connectivity of most areas that constituted the earlier civilizations of the Near East. Slaves from the Black Sea or Thrace could be sold in the port of Athens within a few days of their capture. The short maritime distances involved meant that there was no need for specialized slave ships or slave routes; slaves moved in the same ways and routes as all other commodities.¹⁴ As a result, the price of slaves in classical Athens appears to be far lower in comparison with any other ancient society for which sufficient evidence exists (Rome, Mesopotamia); on average, Athenian slaves costed the equivalent of the annual wages of a skilled craftsman. Given these low prices, it is not surprising that in places like Athens most slaves appear to be first-generation non-Greeks.¹⁵

But what was the role of slave natural reproduction in terms of replenishing slave populations? For certain Greek communities, like Sparta and Crete, the answer is simple: irrespective of the origins of their slave populations, natural reproduction constituted the overwhelming source of their replenishment. As we shall see in the next section, the leisured lifestyle of Spartan and Cretan citizens required large numbers of slaves; but the limited engagement of these societies with Mediterranean connectivity meant that they lacked easy access to networks of supply and the capital required for the constant replenishment through trade: slave reproduction was effectively their only feasible choice. This had important effects on these slave systems: the systemic need for reproduction meant that slave families were relatively stable and that slaves were native inhabitants forming their own communities. This is a major reason why Spartan helots and Cretan woikeis look more similar to medieval serfs than to the standard image of first-generation foreign slaves. The woikeis of Cretan Gortyn had two additional remarkable features. Their slave families had legal consequences: slave children did not belong to the master of the slave mother (the widespread principle of partus ventrem sequitur), but to the master of her senior male relative (father or brother) or of her slave husband. Furthermore, Gortyn allowed mixed marriages between free and slave, something unparalleled in other Greek societies that we know of. The status of the children of these marriages depended on the residence of the couple: if the family resided with the slave husband and was thus under the authority of his master, the children became slaves, while if it resided with the free mother the children were free.¹⁶

For the rest of the Greek world, the significance of slave reproduction is difficult to gauge. In the case of Athens, which is better known, the relatively small numbers of slaves in individual households would have made it difficult to create slave families within the master’s household, a common phenomenon in other societies. Slaves would have often needed to find partners from other households, something that would have created substantial problems in terms of timetables and living arrangements. On the other hand, cities like Athens had substantial populations of slaves who lived and worked on their own; they were more likely to have formed families, and this is often corroborated by the sources. Accordingly, while for many Greek societies like Athens the role of reproduction was secondary compared to trade, the existence of slave families and the role of second-generation slaves was not insignificant.¹⁷

The Experiences of Slaves

Slave experience was shaped by the slaving strategies adopted by their masters. In order to understand these various slaving strategies, it is important to examine the nature and size of the master class. Greek societies were shaped by a fundamental distinction between rich and poor: the rich consisted of all those who were wealthy enough to live without working, while the category of the poor was highly diverse, from the destitute and those barely able to make ends meet to those who lived comfortably, but still had to work alongside their laborers. Leisure was a quintessential aspect of the Greek rich; accordingly, we can take for granted that in most Greek societies the rich were able to afford their lifestyle because they possessed a sufficient number of slaves. There are two main reasons for this: the general absence of institutionalized relations of dependence among the free population and the unwillingness of free people to work for somebody else on a long-term basis, as this was considered akin to slavery. Although free wage laborers were a substantial proportion of some Greek cities like Athens, they would mostly work on short-term contracts for successive employers. Accordingly, long-term workers in households, estates or workshops were almost by definition slaves in the Greek world.

In the case of Athens, where we have sufficient evidence, it appears that rich people possessed on average about 10 slaves. We hear of exceptional cases like the Athenian politician Nicias, who owned a thousand slaves leased to mine operators, and there are examples of owners of workshops that employed 30, 50 or 60 slaves. But it is fairly evident that there was barely any Greek equivalent to the Roman imperial magnates, who possessed urban households with hundreds of slaves, let alone slaves in their rural estates. The orator Demosthenes accused his wealthy opponent Meidias of arrogance for appearing in public spaces accompanied by three slaves¹⁸; this gives a good impression of the relative size of Athenian slaveholdings. We can conclude that while a few very rich people might own tens (and occasionally hundreds) of slaves, most rich people had much smaller slaveholdings.

While slave ownership among the rich is beyond doubt, it is more difficult to assess its extent among the rest of the population. During the archaic period, Sparta and Crete extended the lifestyle of the leisured gentleman to the whole citizen body; as a result, every citizen at Sparta and most citizens in Crete were slave-owners who devoted their lives to warfare, politics and leisure pursuits, because their slaves performed all necessary labor tasks. Sparta and Crete were exceptional; in most other Greek societies, the overwhelming majority of the citizen population had to work for a living. Nevertheless, it is fairly evident that in a rich and powerful society like Athens slave ownership extended to a significant section of working citizens, perhaps one-third of all citizens. Most of these non-elite citizens would own only one or a few slaves. Whether this also applies to other Greek communities, it is impossible to say. Accordingly, while huge slaveholdings, like those attested in Rome or the New World, were quite rare in the Greek world, at the same time slave ownership extended far beyond the elite. The particularly cheap prices of slaves that we mentioned above are probably a major reason for this phenomenon.¹⁹

Households were the key units of social and economic practices in the ancient Greek world. Most Greek slaves were members of relatively small households, consisting of the nuclear family of the master, a few younger or older relatives and a couple of slaves. Accordingly, Greek slavery was deeply shaped by the economic and social strategies of households in terms of acquiring and maintaining property, ensuring a livelihood, producing heirs and transmitting property. A crucial parameter in this respect was the extent to which household heads had free rein to pursue their aims, or were limited by countervailing tendencies and factors. Were household heads allowed to marry, recognize as heirs and bequeath their property to whomever they wanted, or were there rules that imposed, prohibited or prioritized particular courses of action? While most Greek slaves operated within households, an important development of Greek history is the emergence of a new context, in which the economic strategies usually pursued within households were expanded and transformed into large-scale operations employing hundreds or even thousands. By the classical period we see operations like the Athenian workshops employing tens of slaves, while the Athenian mining operations in Laureion used thousands of slaves.²⁰

In what ways were slaves employed? Earlier approaches prioritized the use of slaves in production, and in particular agriculture, as the key criterion of the importance of slavery in a society; other uses, like household service, were considered to be of secondary importance. But this is unnecessarily restrictive. Without modern technological advances that provide fresh water, electricity for lighting and cooking, washing appliances for clothes and dishes and disposable nappies, an enormous quantity of labor was required to perform essential everyday activities, like cutting wood, drawing water and making bread; the various forms of the sexual exploitation of slaves were crucial parameters of the gender and sexual structures of ancient Greek societies; the employment of slaves by Greek states defined their character and activities. At the same time, the economic role of slavery did not take a single form, but a range of diverse forms with very different implications. We should therefore pay equal attention to all the diverse ways in which slaves were employed in ancient Greek societies; the variety of these uses enables us to escape from the structuralist assumptions that have dominated earlier approaches. Many of these uses were compatible with each other and even complementary; but they could also be contradictory and even incompatible. They could therefore generate important stress points and areas of conflict, while also providing slaves with opportunities that would otherwise have been impossible. To this end, we can distinguish between different slaving strategies and examine the full range of strategies that co-existed within a single society.

We can distinguish between five major slaving strategies in the Greek world. A first set of strategies focused on the extraction of labor; within this set, we can further distinguish two subsets: the use of slave labor for maintenance, by employing slaves for the drudgery required for the everyday maintenance of households (cooks, cleaners, personal attendants, nannies), and the employment of slave labor for the production of wealth on rural estates, in workshops and mines. As we have seen, Greek households, even those of the very rich, usually employed only a few slaves. On the other hand, the Athenian mines employed thousands of slaves at their peak, while we know of workshops that employed tens of slaves. Our sources record nothing equivalent for Greek agriculture, but this seems to reflect the fact that even rich Greeks possessed a series of dispersed landholdings, rather than large unified estates. We should therefore expect that such kinds of landholdings required small group of slaves for their cultivation.

In this set of strategies, slaves usually worked in labor processes under the direct control of their masters. Given the unwillingness of free Greeks to be at the constant beck and call of a long-term employer, the practically exclusive use of slaves for household maintenance played a crucial ideological role in Greek societies, by supporting the illusion of citizen equality that was crucial in particular for democracies like Athens.²¹ Given the defining role of leisure for Greek social structures, strategies of labor extraction performed the crucial task of absolving masters and mistresses from the need to work to produce wealth and maintain households. Slavery was also crucial for gender structures: since respectable Greek women were supposed to avoid public spaces and stay indoors, the possession of slaves who could accompany their mistresses allowed mistresses to do things that otherwise would have been unacceptable.

Another set of slaving strategies aimed at revenue extraction rather than labor; in this set masters withdrew from the labor process and used slaves like other possessions and investments which brought revenue, such as real property or loans. We can again distinguish between two subsets. In the first subset masters hired their slaves to other people, who could not afford to buy their own slaves, or had short-term or temporary labor needs that made hiring preferable. In the second subset, masters allowed their slaves to work on their own as cultivators, artisans or traders, on condition that they surrendered part of their earnings. The strategies of revenue extraction were of crucial importance for all forms of Greek slave systems. Spartan citizens were obliged to live in Sparta, but the most fertile part of Spartan territory was Messenia, separated from Sparta by the Taygetos mountain range and difficult to access. As a result, Spartan masters were absentee landowners; this allowed their Messenian helots to effectively operate as dependent peasants, who lived in their own villages, organized the labor process on their own and surrendered only part of the crops to their masters.²² In the highly urbanized societies of the second group of slave systems, independent slaves played major roles as traders and artisans; slave bankers were among the richest inhabitants of Athens. These slaves were relatively independent and were often indistinguishable from the free lower classes; they could often use their hard-won savings in order to enhance their living and ultimately buy their freedom.²³

A third slave strategy concerns gratification: the use of slaves to provide sensory pleasure in all its various forms. Gratification slaving involved musicians, singers, dancers, barbers, masseurs, hairdressers and cooks; but the most important form of gratification concerned sex. The sexual economy of Greek societies was organized on the basis of distinct gender roles. Respectable women could have legitimate access to sex exclusively through marriage; on the contrary, men could legitimately have sex outside marriage, as long as they refrained from having sex with respectable women. But women without honor were fully usable, and slaves constituted the overwhelming majority of women without honor. Furthermore, while girls were commonly married as soon as they reached puberty, men in Greek societies normally deferred marriage until their late twenties or thirties, when they would have received their inheritance and could support their families. As a result, men faced a window of opportunity between puberty in the late teens and marriage in the early thirties in which sexual access took place outside wedlock and was provided by women without honor.

The casual sexual exploitation of slaves by their masters was a ubiquitous feature of Greek societies; at the same time, the high levels of urbanization in the Greek world created the necessary population density for the emergence of brothels; spurred by the widespread availability of slaves and the gender prescriptions of Greek sexual economies, prostitution increased exponentially. The sexual exploitation of slaves fundamentally shaped how gender and sex operated in the Greek world. At the same time, the use of slaves for gratification also led to the creation of relatively stable relations between masters and female slaves (relations between mistresses and male slaves were considered beyond the pale). Concubinage was a common phenomenon. In the Homeric world, masters could recognize the children of their female slaves; this seems to have remained the case in certain Greek societies. But in most Greek societies the intervention of the political community curtailed or prohibited getting heirs through slaves; in Athens, masters could not recognize their slave progeny as legitimate heirs even if they wanted to.²⁴

The fourth set of strategies is the most paradoxical, for it employs slaves for expertise, trust and authority. Slaves were often employed as commercial agents and managers; given the limited scope of Greek legal systems for the delegation of authority among free people, slavery allowed masters to have full control over those that run their business interests. Slaving for expertise was also important for processes of knowledge transmission. In the absence of institutionalized systems of intergenerational training and knowledge transmission, like those of the medieval guilds, buying slaves and training them was a particularly efficient way of creating, maintaining and controlling a specialized workforce. While free trainees could move or become antagonists, trainers of slave experts could control them in a much more guaranteed way.²⁵ Equally remarkable is how Greek states employed public slaves as clerks, bureaucrats and even policemen, like the 300 Scythian archers employed by Athens. The image of public slaves imposing order on citizens was something inconceivable in other slave systems, such as the US South. Public slaves offer a valuable window into Greek politics. In many societies the various groups of the state sector (bureaucrats, judges, policemen, the military) develop their own interests and are able to prioritize them over those of the citizenry as a whole; the use of public slaves allowed Greek states to eschew the development of a powerful bureaucracy, while also allowing states with high levels of citizen participation and magistrate turnover to maintain institutional know-how and continuity.²⁶

The final slaving strategy is that of prestige creation for their masters. This was certainly not unknown in the Greek world: a satirist describes how a man of petty ambition in Athens would buy an exotic Ethiopian slave to flaunt his putative wealth.²⁷ But in contrast to the huge slave retinues that were the necessary accompaniment of elite Romans in public, the role of slaving for prestige creation in the Greek world was highly circumscribed. Greek city-states carefully orchestrated how elite citizens could gain honor and prestige, primarily through acts of public service and munificence; slaving offered private wealth more influence in gaining honor than Greek communities were willing to accept. This is a good example of how slaving strategies allow us to explore the divergent anatomies of different ancient and modern societies.

As the above summary of Greek slaving strategies indicates, the experiences of Greek slaves could differ significantly.²⁸ There was a world of difference between the experiences of mining slaves, slave prostitutes, slave bankers, slave policemen and slave artisans who worked on their own. At the same time, slave experiences were also shaped by wider features of Greek societies, as well as by slave agency. Given that Greek slaves did not belong to a single racial group, it was usually impossible to tell apart slave and free persons by visual means, though there existed certain bodily characteristics that were more common among certain groups of non-Greeks, such as red hair, and could therefore make certain non-Greek slaves easily identifiable. At the same time, Greek cities, in particular from the later classical period onward, contained significant numbers of free immigrants from various parts of the Mediterranean and the Black Sea. Accordingly, e.g., Athenians did not encounter Thracians solely as slaves; the Thracian presence in Athens included free immigrants, freedpersons and slaves. The co-existence of free and enslaved members of the same ethnic and cultural groups had important implications for the shaping of slave identities and ethnicities in the Greek world.²⁹

Another factor that often complicated things was the absence of labor divisions based on status; with the exception of household service and mining, which were exclusively performed by slaves, in all other professions and tasks free and slave laborers worked side by side. It was thus difficult to tell the status of an individual solely on the basis of profession or living standards, as many independent slave artisans or traders lived in conditions identical to those of their free counterparts.³⁰ As one posh ancient author commented, in Athens one should avoid hitting a person because he looked poor, as it was impossible to know whether he was slave or free, and the citizen poor had power because of the democratic political system.³¹ Finally, from the later classical period we see the explosion of associations based on common adherence to a cult, ethnicity or profession. While most members of these associations were free, they could also include freedpersons and even slaves. These mixed associations created networks of solidarity and support that could often play a crucial role in slave strategies for improving their lot.³² As a result of all these factors, passing as free was an important strategy adopted by slaves in the Greek world. In many cases, this was not an attempt to flee from the master, but merely an effort to avoid prejudice and mistreatment from third parties. It is telling in this respect that slaves with “white collar” jobs tended to have standard Greek names shared by the citizens and therefore indistinguishable, while slaves with “blue collar” jobs often bore foreign or ethnic names that made them immediately distinguishable.³³

Exits

There were diverse forms of exits from slavery across the Greek world; they include manumission, either by individual masters for their own reasons or through state intervention for public reasons; and, finally, individual/collective flight or rebellion. Greek communities employed different forms of manumission: in some communities, manumissions were private and informal, requiring little more than a unilateral verbal utterance by the master in the presence of witnesses, or a written will. In other communities we find the phenomenon called “sacral manumission”: the master formally sold the slave to a deity, which paid the master with money provided by the slave on condition that the slave was then manumitted. The origins and meaning of this practice are debated, but it is clear that the deity acted as the middleman that ensured the validity of the contract, as slaves strictly speaking could not contract with their own masters. We also encounter the phenomenon of masters consecrating slaves to deities; such slaves were strictly speaking not free, but the absence of a concrete human master meant that in practice consecrated slaves lived as freedpersons, merely offering their services to a local temple on festival days. While all these forms of Greek manumission were unilateral acts of the masters, we also encounter Greek communities in which communal assent to private manumissions was required; in the most extreme case we know of, that of Sparta, manumission by masters for private reasons was prohibited, and only the state could manumit helots for public purposes.³⁴

How did slaves manage to gain their freedom? In certain cases, masters decided to manumit their slaves free of charge or for a nominal price. This was usually the result of strong interpersonal relations between masters and slaves: manumission was a reward for long and faithful service, or because masters wanted to liberate their slave concubines and their progeny. But in most cases slaves paid for their freedom, often more dearly than simply in monetary terms; one of the most heartbreaking aspects of Greek manumission is the obligation of female slaves to give birth and surrender to their masters a set number of children who would take their place in slavery. The ability of slaves to procure money for their manumission highlights two things: on the one hand, the significance of slaving strategies for revenue, trust and expertise, which created independent slaves who could keep part of their earnings; on the other hand, the significance of networks and communities that involved slaves, freedpersons and free people, who allowed slaves to borrow or pool resources in order to gain their freedom.

What was the status of freedpersons in the Greek world? In the world depicted by Homer, incorporation of freedpersons in the free community appears relatively straightforward: Odysseus promises to reward his faithful slaves with land and wives, presumably indicating their full incorporation in the local community.³⁵ But archaic political processes started to change significantly the nature of community membership in the Greek city-states; gradually, citizenship became a highly codified status defined by the political community, bestowed through specific rituals and often requiring descent from both citizen parents. As a result, from the later archaic period freedpersons in most of the Greek world acquired the status of resident foreigners, if they chose to stay in the same place where they had lived as slaves. Like freeborn resident foreigners, freedpersons could occasionally gain citizenship for major benefactions to the city-state, but that was a very rare phenomenon. It is possible that in some communities of central and northern Greece in the Hellenistic period freedpersons could gain political rights alongside their manumission, though details are unclear.³⁶

A peculiar aspect of the status of Greek freedpersons is the practice of paramone. In a significant number of cases, manumitted slaves were obliged to remain (paramenein) with their former masters for a specified period of time or until the death of the latter; many manumission contracts explicitly state that former masters had the right to punish freedpersons as slaves, or even annul their manumission if they considered them ungrateful or insubordinate, while other contracts prohibit servile punishment and institute panels of arbitrators to settle disputes between former masters and people in paramone. Were people in paramone slave or free? The best answer seems to be that they were free in regard to everybody else, while their status vis-á-vis their former masters depended on the terms of their manumission, ranging from continuous servile subordination to free dependence.

Another important means of exit from slavery concerned the role of the political community. The Greek world was an anarchic geopolitical environment consisting of hundreds of small, medium and large city-states, in which warfare and civil war were ubiquitous phenomena. States could resort to public manumissions in order to enhance their manpower or mitigate a potentially lethal crisis; in 406 BCE the Athenians manumitted and even enfranchised thousands of slaves in order to man a new fleet in a desperate but successful attempt to save their fleet from Spartan blockade. In 86–85 BCE Ephesos in Asia Minor, caught in the middle of the war between King Mithridates and Rome, resorted to manumitting public and private slaves in order to enhance its army and avoid the risk of city betrayal under siege. In wars between different communities or civil conflicts within the same community one or both sides could attempt to strengthen their side and/or destabilize the opposite side by inviting slaves to flee or rebel, promising freedom as a reward. A characteristic example is that of the civil war in Corfu in 427 BCE, when both democrats and oligarchs offered the rural slaves freedom for choosing their side; the majority of the slaves took the side of the democrats, a pattern that was repeated on other occasions as well.³⁷

The co-existence of hundreds of neighboring city-states and the ubiquity of war among them made flight relatively easy, as a different political authority existed just a few miles down the road. There was obviously no guarantee that slaves who flew to a different city would remain free there, rather than being enslaved by a different master, but it was often worth trying; in order to counter that threat, Greek cities often signed bilateral treaties for the mutual return of fugitives. The existence of communities of independent slaves and freedpersons in most urban communities could provide a safe environment for fugitives. Maroon communities were a common phenomenon in many areas of the New World, but we are ill informed about Greek examples. The most famous case is that of Drimacus, who led a maroon community in Chios, ultimately forcing the Chians to reach a modus vivendi, in which Drimacus was recognized as leader of a maroon quasi-state.³⁸

Slave revolts appear to be a rare phenomenon in the Greek world. Among the few examples, the best known is the revolt of the Messenian helots in the 460s BCE, with the rebels ultimately obtaining a truce allowing them to go into exile as free persons. The helots of Messenia were ultimately liberated in 368 BCE, when Thebes defeated Sparta and created an independent Messenian state. Helot revolts profited from the peculiar geopolitical setting of Sparta; none of its neighbors had a similar form of slave system, and they were willing therefore to encourage helot revolt without fearing the Spartans doing the same.

The Peloponnesian War (431–404 BCE) offers a more common example of how slaves gained their freedom in mass numbers. In the early phase of the war, the Athenians occupied a fort in Spartan territory and encouraged the helots to flee; many did, and the Spartan fear of a wider revolt led them to agree to a humiliating treaty that ended the first part of the war. But the Spartans also decided to use thousands of helots as soldiers by promising them freedom, and these helot soldiers scored some of the most decisive Spartan victories. When the Spartans later in the war also occupied a fort in Athenian territory, thousands of Athenian slaves flew to them, although most of them did not gain their freedom in this way, but merely changed masters. This mass flight seriously debilitated Athenian fortunes; at the same time, thousands of Athenian slaves gained freedom and even citizenship by joining the Athenian navy, while many others joined the movement that restored Athenian democracy in the aftermath of the war. Athenian slaves and Spartan helots did not act in defense of collective slave interests; different groups acted in different ways, according to circumstances and choices. But slave agency played a crucial role in the outcome of the war, and thousands of slaves gained their freedom in one way or another.

Slavery was a defining phenomenon of the history of Greek city-states. The diverse slaving strategies that co-existed in most Greek societies meant that slavery could be used for a variety of purposes, partly complementary, and partly contradictory. While Greek elites depended on slaves, substantial middling strata of Greek societies were also slaveholders; Greek civic institutions and the material culture of Greek consumer societies were fundamentally shaped by the ubiquity of slavery; the connectivity and market expansion of the first millennium BCE had slavery in its very core. At the same time, the processes in which slavery was inscribed could lead to very divergent outcomes among Greek slave systems. Studying the history of these diverse systems is a potent means of understanding the historicity of slavery as a global phenomenon.

Notes

• 1. Moses I. Finley, Ancient Slavery and Modern Ideology (London: Chatto and Windus, 1980).

• 2. This is still the dominant approach in Keith Bradley and Paul Cartledge, eds. The Cambridge World History of Slavery 1: The Ancient Mediterranean World (Cambridge: Cambridge University Press, 2011).

• 3. David M. Lewis, Greek Slave Systems in Their Eastern Mediterranean Context, c. 800–146 BC (Oxford: Oxford University Press, 2018); Kostas Vlassopoulos, Historicising Ancient Slavery (Edinburgh: Edinburgh University Press, 2021).

• 4. For the concept of slaving and no-slaving zones, see Jeffrey Fynn-Paul, “Empire, Monotheism and Slavery in the Greater Mediterranean Region from Antiquity to the Early Modern Era,” Past & Present 205, no. 1 (2009): 3–40.

• 5. Thucydides, History, 8.40.2.

• 6. Nemenja Vujčić, “A Numbers Game: The Size of the Slave Population in Classical Athens,” Zbornik: Journal of Classical Studies 23 (2021): 87–111.

• 7. Edward M. Harris, “Did Solon Abolish Debt-bondage?” Classical Quarterly 52, no. 2 (2002): 415–30.

• 8. Vincent Gabrielsen, “Warfare, Statehood and Piracy in the Greek World,” in Seeraub im Mittelmeerraum: Piraterie, Korsarentum und maritime Gewalt von der Antike bis zur Neuzeit, ed. N. Jaspert and S. Colditz (Paderborn: Ferdinand Schöningh/Wilhelm Fink Verlag, 2013), 131–53.

• 9. Poetae Melici Graeci no. 909.

• 10. David M. Lewis, “Piracy and Slave Trading in Action in Classical and Hellenistic Greece,” Mare Nostrum 10, no. 2 (2019): 79–108.

• 11. Hans van Wees, “Conquerors and Serfs: Wars of Conquest and Forced Labor in Archaic Greece,” in Helots and Their Masters in Laconia and Messenia: Histories, Ideologies, Structures, eds. N. Luraghi and S. Alcock (Cambridge, MA: Center for Hellenic Studies, 2003), 33–80.

• 12. Pierre Ducrey, Le traitement des prisonniers de guerre dans la Grèce antique, des origines à la conquête romaine (Paris: De Boccard, 1968); Anne Bielman, Retour à la liberté: Libération et sauvetage des prisonniers en Grèce ancienne. Recueil d’inscriptions honorant des sauveteurs et analyse critique (Athens: Ecole Française d’Athènes, 1994).

• 13. Xenophon, Anabasis, 7.3–4.

• 14. David M. Lewis, “The Market for Slaves in the Fifth-and Fourth-Century Aegean,” in The Ancient Greek Economy: Markets, Households and City-States, eds. E. Harris, D. M. Lewis and M. Woolmer (Cambridge: Cambridge University Press, 2015), 316–36.

• 15. Walter Scheidel, “Real Slave Prices and the Relative Cost of Slave Labor in the Greco-Roman World,” Ancient Society 35 (2005): 1–17.

• 16. Kostas Vlassopoulos, “Historicising the Closed City,” in La cité interconnectée: transferts et réseaux institutionnels, religieux et culturels aux époques hellénistique et impériale, eds. M. Dana and I. Savalli-Lestrade (Bordeaux: Ausonius, 2018), 43–57.

• 17. Winfried Schmitz, “Sklavenfamilien im archaischen und klassischen Griechenland,” in Kindersklaven-Sklavenkinder: Schicksale zwischen Zuneigung und Ausbeutung in der Antike und im interkulturellen Vergleich, ed. H. Heinen (Stuttgart: Franz Steiner Verlag, 2012), 63–102.

• 18. Demosthenes, Against Meidias, 157–8.

• 19. Jason D. Porter, “Slavery and Athens’ Economic Efflorescence”, Mare Nostrum 10, no. 2 (2019): 25–50.

• 20. Siegfried Lauffer, Die Bergwerkssklaven von Laureion, I-II (Wiesbaden: Franz Steiner Verlag, 1956).

• 21. Robin Osborne, “The Economics and Politics of Slavery at Athens,” in The Greek World, ed. A. Powell (London: Routledge, 1995), 41–57.

• 22. Stephen Hodkinson, “Spartiates, Helots and the Direction of the Agrarian Economy: Towards an Understanding of Helotage in Comparative Perspective,” in Slave Systems Ancient and Modern, eds. E. Dal Lago and C. Katsari (Cambridge: Cambridge University Press, 2008), 285–320.

• 23. Jason D. Porter, “The Apophora and the ‘Leasing’ of Property to Slaves and Manumitted Slaves in Classical Athens,” Historia 70, no. 2 (2021): 185–205.

• 24. Edward E. Cohen, “Sexual Abuse and Sexual Rights: Slaves’ Erotic Experience at Athens and Rome,” in A Companion to Greek and Roman Sexualities, ed. T. K. Hubbard (Chichester: Blackwell, 2013), 184–98.

• 25. Tracy E. Rihll, “Skilled Slaves and the Economy: The Silver Mines of the Laurion,” in Antike Sklaverei, Rückblick und Ausblick: Neue Beiträge zur Forschungsgeschichte und zur Erschließung der archäologischen Zeugnisse, ed. H. Heinen (Stuttgart: Franz Steiner Verlag, 2010), 203–20.

• 26. Paulin Ismard, Democracy’s Slaves: A Political History of Ancient Greece (Cambridge, MA: Harvard University Press, 2017).

• 27. Theophrastus, Characters, 21.4–5.

• 28. Hans Klees, Sklavenleben im klassischen Griechenland (Stuttgart: Steiner Verlag, 1998).

• 29. Peter Hunt, “Trojan Slaves in Classical Athens: Ethnic Identity among Athenian Slaves,” in Communities and Networks in the Ancient Greek World, eds. C. Taylor and K. Vlassopoulos (Oxford: Oxford University Press, 2015), 129–54.

• 30. Kostas Vlassopoulos, “Free Spaces: Identity, Experience and Democracy in Classical Athens,” Classical Quarterly 57 (2007): 33–52.

• 31. Pseudo-Xenophon, Constitution of the Athenians, 1.10.

• 32. Stéphanie Maillot, “Associations dites d’étrangers, clientèles et groups de travail à l’époque hellénistique,” in Statuts personnels et main-d’œuvre en Méditerranée hellénistique, eds. S. Maillot and J. Zurbach (Clermont Ferrand: Presses Universitaires Blaise-Pascal, 2021), 285–313.

• 33. Kostas Vlassopoulos, “Athenian Slave Names and Athenian Social History,” Zeitschrift für Papyrologie und Epigraphik 175 (2010): 113–144.

• 34. Rachel Zelnick-Abramovitz, Not Wholly Free: The Concept of Manumission and the Status of Manumitted Slaves in the Ancient Greek World (Leiden: Brill, 2005).

• 35. Homer, Odyssey, 21.209–16.

• 36. Rachel Zelnick-Abramovitz, “Freed Slaves, Their Status and State Control in Ancient Greece,” European Review of History / Revue européenne d’histoire 16, no. 3 (2009): 303–18.

• 37. Karl-Wilhelm Welwei, Unfreie im antiken Kriegsdienst, I-III (Stuttgart: Franz Steiner Verlag, 1974–1988).

• 38. Sara Forsdyke, Slaves Tell Tales and Other Episodes in the Politics of Popular Culture in Ancient Greece (Princeton: Princeton University Press, 2012), 37–89.